Mamba chunk scan with segsum
Mamba-2 chunk-based parallel scan with segment sum computation for intra-chunk and inter-chunk state propagation. This is the core parallel scanning algorithm that enables efficient SSM computation by breaking sequences into chunks (chunk_size=256), computing intra-chunk attention-like patterns using segment sums, maintaining inter-chunk state recurrence with exponential decay, and combining diagonal (intra-chunk) and off-diagonal (inter-chunk) contributions.
Reported evidence · last observed 2026-07-10. Reported by source; not independently reproduced. No public source. License unknown.
Current records
Workloadsuite of 16 cases · mean latencyhead_dim = 64 · n_groups = 1 · num_heads = 16 · chunk_size = 256 · hidden_out = 1024 · state_size = 25616 cases
Not measured on H100 for this workload. Challenges →
148.5µs93% SOL1.00×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
260.8µs92% SOL1.25×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
362.5µs92% SOL1.29×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
464.9µs91% SOL1.34×Reported · license unknown · no source2026-08-13
Measured exactly what you asked. Reported by source; not independently reproduced.
568.0µs91% SOL1.40×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
673.9µs90% SOL1.52×Reported · license unknown · no source2026-07-21
Measured exactly what you asked. Reported by source; not independently reproduced.
784.6µs89% SOL1.74×Reported · license unknown · no source2026-07-17
Measured exactly what you asked. Reported by source; not independently reproduced.
899.0µs87% SOL2.04×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
999.1µs87% SOL2.04×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
1099.1µs87% SOL2.04×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
11107.2µs86% SOL2.21×Reported · license unknown · no source2026-07-26
Measured exactly what you asked. Reported by source; not independently reproduced.
12473.4µs58% SOL9.75×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.